基于轨迹数据的特定用户行为分析方法  

Specific User Behavior Analysis Method Based on Trajectory Data

作　　者：高翔 GAO Xiang(School of Digital Media,Lanzhou University of Arts and Science,Lanzhou 730000,China)

机构地区：[1]兰州文理学院数字媒体学院,甘肃兰州730000

出　　处：《兰州文理学院学报(自然科学版)》2025年第2期54-58,共5页Journal of Lanzhou University of Arts and Science(Natural Sciences)

基　　金：兰州市科技计划项目(2022-2-74)。

摘　　要：为了实现对管控人员的监督,提高管控效率,提出基于轨迹数据的特定用户行为分析方法.通过智能设备对管控人员的位置和状态信息进行采集,然后根据坐标点之间相关系数挖掘轨迹中的停留点,根据距离和时间阈值筛选停留区域.停留区域相互连接形成管控人员的行动轨迹,为了便于管控人员行为分析,将长轨迹进行二次切割,建立DBSCAN聚类模型,挖掘用户频繁路径并减少其缺失,从而为特定人群的监管提供支持.In order to realize the supervision of management personnel and improve the efficiency of management and control,a specific user behavior analysis method based on trajectory data is proposed.The location and status information of the control personnel is collected through intelligent equipment,and then the stay points in the trajectory are mined based on the correlation coefficient between coordinate points,and then the distance and time thresholds are set to filter the stay areas among the stay points.The stay areas are connected to each other to form the action track of the controller.In order to facilitate the behavior analysis of the controller,the long track is cut twice and the DBSCAN clustering model is established to mine the frequent paths of users and reduce the absence of frequent paths,thus providing support for the supervision of specific groups.

关 键 词：轨迹分析 聚类算法 用户行为分析 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]